Video encoding/decoding method and apparatus for improving merge mode

Abstract

A method and a device for encoding/decoding a video are provided. A method for decoding a video according to the present disclosure includes determining a first mode based on at least one of a combined inter/intra prediction (CIIP) mode, a geometric partitioning mode, or a merge mode with motion vector difference. The method for decoding the video also includes determining a motion vector from a merging candidate list. The method for decoding the video also includes generating a prediction block of a current block based on the first mode and the motion vector.

What is claimed is: 
     
         1 . A method of decoding video, the method comprising:
 determining a first mode based on at least one of a combined inter/intra prediction (CIIP) mode, a geometric partitioning mode, or a merge mode with motion vector difference;   determining a motion vector from a merging candidate list; and   generating a prediction block of a current block based on the first mode and the motion vector.   
     
     
         2 . The method of  claim 1 , wherein when the first mode is a combined mode of the CIIP mode and the geometric partitioning mode, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the CIIP mode and the geometric partitioning mode, generating the prediction block of the current block comprises:   generating a final inter-prediction block for the current block based on the two partitioned regions and the motion vectors for the two partitioned regions;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the final inter-prediction block and the intra-prediction block.   
     
     
         3 . The method of  claim 2 , wherein generating the final inter-prediction block for the current block comprises:
 generating inter-prediction blocks respectively for the two partitioned regions by using the motion vectors for the two partitioned regions; and   generating the final inter-prediction block for the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions.   
     
     
         4 . The method of  claim 1 , wherein when the first mode is a combined mode of the CIIP mode and the merge mode with motion vector difference, generating the prediction block of the current block comprises:
 generating an inter-prediction block for the current block based on a refined motion vector;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the inter-prediction block and the intra-prediction block,   wherein the refined motion vector is generated based on the motion vector and motion vector difference information.   
     
     
         5 . The method of  claim 1 , wherein when the first mode is a combined mode of the geometric partitioning mode and the merge mode with motion vector difference, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the geometric partitioning mode and the merge mode with motion vector difference, generating the prediction block of the current block comprises:   generating inter-prediction blocks respectively for the two partitioned regions by using refined motion vectors; and   generating the prediction block of the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions,   wherein the refined motion vectors are generated based on the motion vectors for the two partitioned regions and motion vector difference information.   
     
     
         6 . The method of  claim 1 , wherein when the first mode is a combined mode of the CIIP mode, the geometric partitioning mode, and the merge mode with motion vector difference, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the CIIP mode, the geometric partitioning mode, and the merge mode with motion vector difference, generating the prediction block of the current block comprises:   generating a final inter-prediction block for the current block based on the two partitioned regions and refined motion vectors for the two partitioned regions;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the final inter-prediction block and the intra-prediction block,   wherein the refined motion vectors are generated based on the motion vectors for the two partitioned regions and motion vector difference information.   
     
     
         7 . The method of  claim 6 , wherein generating the final inter-prediction block for the current block comprises:
 generating inter-prediction blocks respectively for the two partitioned regions by using the refined motion vectors for the two partitioned regions; and   generating the final inter-prediction block for the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions.   
     
     
         8 . A method of encoding video, the method comprising:
 determining a first mode based on at least one of a combined inter/intra prediction (CIIP) mode, a geometric partitioning mode, or a merge mode with motion vector difference;   determining a motion vector from a merging candidate list; and   generating a prediction block of a current block based on the first mode and the motion vector.   
     
     
         9 . The method of  claim 8 , wherein when the first mode is a combined mode of the CIIP mode and the geometric partitioning mode, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the CIIP mode and the geometric partitioning mode, generating the prediction block of the current block comprises:   generating a final inter-prediction block for the current block based on the two partitioned regions and the motion vectors for the two partitioned regions;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the final inter-prediction block and the intra-prediction block.   
     
     
         10 . The method of  claim 9 , wherein generating the final inter-prediction block for the current block comprises:
 generating inter-prediction blocks respectively for the two partitioned regions by using the motion vectors for the two partitioned regions; and   generating the final inter-prediction block for the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions.   
     
     
         11 . The method of  claim 8 , wherein when the first mode is a combined mode of the CIIP mode and the merge mode with motion vector difference, generating the prediction block of the current block comprises:
 generating an inter-prediction block for the current block based on a refined motion vector;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the inter-prediction block and the intra-prediction block,   wherein the refined motion vector is generated based on the motion vector and motion vector difference information.   
     
     
         12 . The method of  claim 8 , wherein when the first mode is a combined mode of the geometric partitioning mode and the merge mode with motion vector difference, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the geometric partitioning mode and the merge mode with motion vector difference, generating the prediction block of the current block comprises:   generating inter-prediction blocks respectively for the two partitioned regions by using refined motion vectors; and   generating the prediction block of the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions,   wherein the refined motion vectors are generated based on the motion vectors for the two partitioned regions and motion vector difference information.   
     
     
         13 . The method of  claim 8 , wherein when the first mode is a combined mode of the CIIP mode, the geometric partitioning mode, and the merge mode with motion vector difference, determining the motion vector from the merging candidate list comprises:
 determining motion vectors respectively for two partitioned regions from the merging candidate list, and   wherein when the first mode is the combined mode of the CIIP mode, the geometric partitioning mode, and the merge mode with motion vector difference, generating the prediction block of the current block comprises:   generating a final inter-prediction block for the current block based on the two partitioned regions and refined motion vectors for the two partitioned regions;   generating an intra-prediction block for the current block based on at least one or more reference pixels adjacent to the current block; and   generating the prediction block of the current block by weighted averaging the final inter-prediction block and the intra-prediction block,   wherein the refined motion vectors are generated based on the motion vectors for the two partitioned regions and motion vector difference information.   
     
     
         14 . The method of  claim 13 , wherein generating the final inter-prediction block for the current block comprises:
 generating inter-prediction blocks respectively for the two partitioned regions by using the refined motion vectors for the two partitioned regions; and   generating the final inter-prediction block for the current block by combining or weighted averaging the inter-prediction blocks for the two partitioned regions.   
     
     
         15 . A method for transmitting a bitstream associated with video data, the method comprising:
 generating the bitsream by encoding the video data; and   transmitting the bitstream to a video decoding apparatus,   wherein generating the bitstream comprises:   determining a first mode based on at least one of a combined inter/intra prediction (CIIP) mode, a geometric partitioning mode, or a merge mode with motion vector difference;   determining a motion vector from a merging candidate list; and   generating a prediction block of a current block based on the first mode and the motion vector.